https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95784
--- Comment #4 from Gabriel Ravier <gabravier at gmail dot com> --- I've tried to benchmark this on my "Intel(R) Core(TM) i5-4310U CPU @ 2.00GHz" (this is from /proc/cpuinfo) and I'm getting some really weird results, so if you want to try to assist me in benchmarking (which should determine which version really is faster) that'd be welcome. I've tried to benchmark this with the attached test.S source file and got these results : $ gcc test.S -O3 -ggdb3 -DGCC_VERSION && time ./a.out && gcc test.S -O3 -ggdb3 -DLLVM_VERSION && time ./a.out && gcc test.S -O3 -ggdb3 && time ./a.out real 0m2.210s # GCC version user 0m2.200s sys 0m0.001s real 0m1.825s # LLVM version is faster as I expected user 0m1.817s sys 0m0.001s real 0m2.080s # Doing nothing is somehow slower than the LLVM version ? user 0m2.071s sys 0m0.001s
